Welcome to my Technical Design blog, coinciding with my study year with Noroff as part of the Technical Design with CAD 2D/3D course. This blog with contain a day to day reflective journal on the various exercises, tasks and objectives throughout the course.
This task was to visualise an existing house designed by the renowned American architect, Louis Kahn. The house is the Esherick House, one of his best known designs. The empty model is provided in 3DS max format ready for the addition of furniture, materials and lighting. This was an exercise in texturing, lighting and animation rendering, with some post-production editing in Photoshop.
I admit my first reactions when receiving the brief were a little panicked! This was a large task, with multiple submission elements to have control over in the space of only 5 days. However, what a fantastic project! I have really found myself in my element, getting to intimately know a classic house by a great architect and attempt to produce renderings which do it justice.
The model that was provided was detailed and coherently organised. It was a benefit to see how such a model was put together and how the extra detail helps in achieving great results. Individual floor boards and other timber had been modelled in detail, leaving me with the task of mapping suitable textures
This phase was important, and my experience so far has taught me that it is wise to start with investigating camera angles. I knew therefore very early on where my focus in materials and objects would be.
Researching the house also gave me a story. I became aware of the original client, and that she was a fanatical reader. This is the reason for the double height bookcase, and numerous reading crooks throughout the house. Plenty of well lit spaces to crouch in with a good book. With this information I already had inspiration for how my final images might look.
The most useful page I discovered was this one, documenting recent owners and their mission to update the house for their own living requirements. It also had some fantastic photographs. I used these to set up camera angles, influence the compositions and returned regularly to reference lighting and materials throughout. However, once set up, I was able to choose some of my own angles but it was a useful study exercise to attempt to replicate these photographs.
The key materials to apply were the beige concrete, and Apitong wood. Kahn’s design expresses these materials in simple rectilinear forms, and I see immediately that their textures are fundamental to the quality of the spaces. For the concrete, I made my own seamless texture using photoshop from the images of the house. The variation in the timber was harder to achieve, but I reached something satisfactory by using an image from textures.com. Different materials were created for floors, exterior and interior wood. Unfortunately it was not a simple task to separate exterior/interior, as the model was built with some geometry facing on both sides. Separating the geometry would have proved to be time consuming, so instead I opted for a texture that could be used in both situations.
I experimented with numerous HDRI maps, and two are used in the final images. In hindsight, the bright daylight worked better and gave better lighting results. For the interior shots, an invisible booster light was applied for better daylighting. With the more overcast images, it was really the reflections of the trees in the windows which were the reason for choosing that particular map at that time.
Model handling and rendering
All furnishing was selected and downloaded from 3Dwarehouse, selected carefully for polygon counts, imported first into a seperate Max file before merging into the file. This gave me the opportunity to control the clutter in the model and hold rendering times to a workable level. This was much more successful than the previous submission, where the model became too heavy to work with.
I also switched to Arnold instead of ART in this project, which may also have helped rendering times. I am only beginning to see the benefits, and to make use of the special lighting and material options which make it a more powerful engine. I see however that it has greater potential than ART and than I am currently able to harness.
New modelling techniques
The mantlepiece was a unique feature which I decided to try my hand at creating using subdivision surface modelling, and mesh editing using soft selection. The result was less than perfect in its subdivisions, but achieved the desired result of the curved form.
To get the brickwork on the patio to align correctly, I had to learn how to apply a texture to a path using the UVW Unwrap modifier using this video.
Two animations were designed and rendered out , focusing on the double height living room space.
An initial walkthrough of the entire house proved to be too complicated and too fast to show detail in a 30 second animation. Instead I chose to focus on the the living room showing various angles and highlighting various points of interest:
Another video shows the Alexander Calder mobile. This was an experiment I started in the previous week, developing an understanding of hierarchical animation. Each of the arms of the mobile has its own rotation, but hangs from another arm which also rotates, giving infinite possibilities for how the mobile may configure itself with the movement of air currents. The model was found on 3Dwarehouse, unanimated. However, I see that it is likely not a true copy of an original as the Calder mobiles were beautifully produced without any colliding parts, and unfortunately I see that some of these geometries actually come into conflict in the model on occasion. However, I am very happy with the effect that animating the mobile has on the space, and the animation.
I already have a fairly good understanding of Photoshop, which I have used for many years already. However, it has been useful to find time to refine techniques and find new methods. I still wish to develop my understanding of colour balancing and compositional techniques, amongst other things.
One of my problems with this task was finding a suitable figure to place on the Chaise Longue by the window, ideally reading a book, as to fit my narrative for the image. A near suitable candidate was found, but not perfectly aligned, and with legs bent much higher than wished. Here she is in the original format:
As you see in the final interior image, I managed to position her at a more convincing angle for the furniture, after flipping the image horizontally. This took some time, and numerous attempts before I stumbled upon the Puppet Warp tool which enabled me to stretch out her legs to the desired angle. After some final tweaking with the liquify filter, some shadowing and colour matching I achieved a result I felt was convincing. I feel it was worth the perseverance: Instead of looking down into an empty room, we are suddenly peeking in on a private moment within a home. After a week of gradually tweaking this composition, with just this one adjustment the whole image suddenly took a new meaning – the one I was intending from the start!
This recommended blog was worth reading, and inspiring. I especially took from it some tips about focus and blurring for this exercise. The shot focusing on the bookcase especially benefited from blurring out the foreground shifting focus to the main subject in this case. I realise there are techniques for this within the render engine, which I will hope to investigate further.
This new module turns to Architectural Visualisation, or #ArchViz as I see it has become these days. Seems like social media is having an effect here too, but not least for the amount of inspirational images appearing under similar hashtags daily.
Firstly, I have a confession… this is not my first time using 3DSMax!
I last used this software to produce images for my BArch submission in 2009. Notable projects I used this for can be found here:
- Chetham’s School of Music
- Museo Diocesano
However, that was some considerable time ago. There are many advancements, and some considerable changes to 3DSMax, notably with materials and rendering. I have also forgotten much of how I produced this earlier work, in what I find is a complicated and sometimes frustrating piece of software. This part is true now as it was then!
Instead of using this project as an introductory course as it might have been intended, I have used my time to relearn the software and develop a deeper understanding of certain aspects of the visualisation process.
The assignment brief presented me with an empty circular room, with access from the rear projecting towards a 270° panorama window.
First of all, I imported the model, set up a couple of overview cameras and experimented with a number of HDRI images downloaded from the website https://hdrihaven.com/.
A Pinterest board was created for circular interiors, used to gather inspiration. From there, I sketched out a concept based on the idea of a circular sunken ‘conversation pit’, with a central chandelier. Other elements would effectively be placed in orbit around the chandelier. The hanging mobiles of Alexander Calder were an additional inspiration. Hanging chairs and a suspended fireplace were placed around the outside of the space to increase the sense of suspension, and so not to interrupt the view outwards. The movement of the Calder mobile, hanging chairs and crackling fire would also help give an extra dynamic to the space in reality, though how much of that could translate to a still image would be limited.
The idea for a private entertainment space on the seafront with a sunset view was therefore conceived, and the sketch below helped form a plan for the following work.
Day 1 – Geometry
One of the most difficult tasks was finding available models to realise the design concept. My final image uses a mixture of models imported from manufacturers resources, and other models found at 3D Warehouse. Other elements, such as raised flooring, carpentry, and pendant lamps, have been modelled directly in 3DSMax.
The greatest challenge was finding a ready made sofa to fit the scale of the room. In retrospect, it would have been best to model this in Max and keep the polygon count down. Concerned I would use all of my time modelling a sofa, what I did was to import a generic model I found on 3D warehouse, and scale this up slightly to the desired size. Unfortunately, I lost track of the relative scale and it wasn’t until later I realised this sofa was gigantic in real world size! The problem was partly solved by duplicating cushions to create a realistic depth, but more unfortunately more cushions meant more polygons. Again, something which would have been better to model rather than import. In the final result there is unfortunately a slight (if subtly disguised!) mismatch between sizes of the sofa and coffee tables. In future, I will be sure to plan a space paying more attention to scale, and not to rely on Max in the design phase.
The test render at the end of this first phase looked like this:
Day 2 – Materials
The second day of work was focused towards texturing the scene. Each material was previewed in session using active render, with a higher exposure value than seen below from the overview camera. This is the test render taken at the end of that day, without lighting:
The key materials to introduce here were wooden textures for the floor (oak) and carpentry (walnut). I invested good time here in getting a desired amount of reflection and texture. The grey upholstery texture was also important, created using a blend of bitmaps and noise texturing to create variation in tone. This was based upon the manufacturers own catalogue. The result of that was pleasing enough that I decided to extend the overview to include an additional reading area in the bottom left corner, so that could be seen more close up.
The rug is a great visual feature in the final image. This was imported directly with a texture, using a model from 3D Warehouse. However, this did make the model a little heavy and slower to render. I combatted that by organising the rug (and all other respective models) in layers which could easily be turned on and off when not required. Again, I could have had better control here with what I was importing.
I would have liked to include fur textures in this model, but this will take some further mastering of material editing.
Day 3 – lighting
In reality, this was not a full day or a task totally separated from the others. But the final day’s work was focused upon getting a satisfactory balance for the maximum visual impact.
After adding all of the lights I required, I saw that the image had lost some of the visual impact from some of the earlier test renders. Using the light lister and exposure controls, I adjusted it back to where I thought it was best to capture that magical moment at sunset where there is balance between natural and artificial lighting. Experiments to reduce the amount of blue hue given off by sky image were conducted, but without satisfactory results. This I understand could be adjusted in Photoshop if desired, but for the purpose of this submission I have only included renderings direct from 3DSMax.
A test render with daylighting:
After adding artificial lighting, but now becoming overlit for the sunset to have any visual impact:
Playing with fire
The material I am perhaps most satisfied with was the flame in the fireplace. With help from this tutorial, I saw how it was possible to create this effect. The warmth radiating from the fireplace is a key element in creating the sense of warmth and cosiness in the space.
The flame geometry comprises two planes, converted to meshes and deformed. The flame texture comprises two maps; one for the base colour and a black/white cutout map created in photoshop by removing saturation, and strengthening the brightness/contrast. The material was then given an emission value with a red colour so that the fire would emit warm light into the room.
A similar process was used to create a flame for the candles in the room, with a much smaller emission value.
The most obvious problem with this model is that the polygon count is high for the hardware I am using. At the point reached, working with this model is now difficult and slow.
The problem was circumnavigated by using layers, and turning off heavy objects when not needed. However, this is not ideal and obtaining satisfactory quality final renderings also takes significantly more time.
This model totalled a polygon count of 1389636.
As an experiment I set the main image to render for the longest time I could. I stopped it after 18 hours and 3 minutes. For a 1920×1080 image the quality reached 31.1dB after 4572 iterations.
Arnold vs ART
Initially, I thought to start learning how to use Arnold. Whilst he sounded very friendly, the experience I had from the start was anything but! For this project I decided to stick with ART to keep things simple whilst I gain a better understanding of how to combine different maps and settings. The material editor has changed significantly since earlier versions of Max, and one I spent considerable time getting to learn in tutorials over the past two weeks.
VRay is unfortunately out of my reach just now, though I understand it is used widely in architectural visualisation. I am guessing many of the concepts will be similar in using Arnold, so perhaps VRay is not worth investing in just yet.
Here are the final renders, after some minimal photoshop tweaking and enhancement:
With Easter plans cancelled by Corona-lockdown and staying indoors generally being the most acceptable thing to do right now, I’ve taken the opportunity to use some extra time getting to know Revit (as well as some long walks in the occasional spring sunshine, and a couple of other lockdown projects!).
It’s only 3 weeks since I started learning Revit as part of this course. I am building on some prior understanding of ArchiCAD / 3DSMax and am not therefore a total novice with BIM or visualisation. However, there is much I wish to learn! So far with Revit I’m delightfully surprised at how easy it is to create fairly decent looking renderings from a simple model.
This task was to draw up a Revit model using the plan above. This project actually began serving as a warm-up to the previous house I posted. I came back to it this week to do a little more experimentation with the various tools, which I look forward to learning more about in detail in the coming weeks.
a house rooted in a typical early Scandinavian modernist style, perhaps inspired by the work of Alvar Aalto.
The original house drawings didn’t have any dates, but from the types of spaces and materials indicated it seems like a house rooted in a typical early Scandinavian modernist style, perhaps inspired by the work of Alvar Aalto. With that in mind I decided to include a little nod to Le Corbusier in some of the interior details, but again it is not exhaustive in terms of refinement of the design. Most of the blocks used, including the Le Corbusier sofa, are standards shipped with the software.
There were a couple of items however I used to experiment with to create my first family groups. The chimney stacks are a custom form, created to mimic the original elevation drawings. I also attempted a parametric picture frame, with customisable dimensions and material. The thought here was to make it simple to reuse the same family component throughout the project, should further various sized picture frames be required. The picture itself was created applying a jpg decal of the Modular Man into the imported frame model.
The fireplace was a butchering of a simple downloaded model, with adapted sizes and material. It most likely wouldn’t be permitted by building regulations in a new-build house, but I treated this as if it was an existing structure, from the 1930s for example.
This project is really just a trial of some of the basic features of Revit, without going exhaustively into the design or detailing. The interior shot was created in a couple of hours, with very minimal furnishing. I am quite pleased however to present my first (purely Revit produced, no post production) renderings produced using the software!
Mastering the use of Revit and BIM was one of the reasons I was attracted to this course. I have been therefore super-motivated to get stuck in to this task and start getting to grips with this program!
At the end of the first two weeks, I am a little struck with the complexity of the software and the range of configurations. To me this is a totally new way of constructing a 3D model. Creating geometry in AutoCAD and Inventor seems to be much more intuitive, and I see this can be imported into a Revit family if desired. However, I can see the usefulness of the way the model is constructed in BIM as opposed to regular drawing.
The 2D drawing output from my model is not quite as refined as I would like yet. Since the course material so far has only covered basic modelling, I have not attempted to refine these 2D drawings in any way other than to make sure the geometry produced by the model is correct. I see there are display and annotation settings which can do this, which I hope to go into at a later point.
Similarly, rendering and 3D imaging settings have also been ignored with the hope of learning more later. I hope I will look back on this week’s blog entry in the future and consider this model raw and basic! At this point, getting to grips with this seems anything but simple. ‘My first Revit project’ is therefore a nod to that this really is a kind of ABC-123 type introduction on my part, and about learning to crawl through the jungle of modelling options available in this software!
Create a BIM model of the house below from a set of 2D drawings using BIM objects from standard suppliers.
Full dimensioned 2D drawings, including plans, sections and elevations have been made available. Additionally, specific doors are to be used from the NorDan and Swedoor product ranges for which models from BIMObject are available for use in the model.
In addition, a wooden staircase is to be drawn using the built-in Revit stair tool.
As extra challenges, the following may be included:
- Custom wall type, as real as possible using external sources
- Custom floor type, as real as possible using external sources
Glava insulation was used as a reference point for giving a suitable wall build-up. The studwork dimension on the provided drawings was 148mm, which I combined with a 48mm outer wall panel, and a total of 50mm for external and internal finishes to give a total of 246mm.
The same principle was applied to floors as to the walls. Both the ground floor and dividing floor between level 1 and 2 were taken from the Glava manual, using the overall depth given on the original section drawing (and disregarding current U-value requirements).
To create the ringwall foundation, I began with the Structural Foundation: Wall tool. A problem which came up here was that despite setting this to 200mm blockwork, the software wanted to extend the thickness of the ring wall to match the wall above, whereas this should be level with the core above.
I later changed the ringwall to a standard wall, with a thickness of 200mm in concrete blockwork, ignoring any detail for damp protection at this stage. I also included a foundation under this wall. The problem here is that the cladding does not overlap the joint as it should.
It seems like Revit has a lot of functions which aren’t necessarily obvious or easy to find. Extending this cladding downwards was straightforward in the end, but not before several hours of searching through and watching various tutorials. This one eventually explained how to get the cladding to overlap by unlocking the bottom of the preview in the ‘Modify Vertical Structure’ function in the ‘Edit Assembly’ view. This section could then be manually edited in section view.
One of the most useful features I have learned so far is how easy it is to create cutaway 3D views illustrating various aspects of the construction, using the Selection Box tool.
This stair tool created some headaches in generating the desired form! I had to adjust the riser and tread dimensions in order to be able to position this to fit with the door and floor openings. I wanted also to make sure it complied with PBL for stair dimensions, which states a minimum tread of 250mm. Interestingly, unlike British regulations, there is no maximum riser height (220mm for private dwellings). With 15 risers, the height worked out at being 197mm each and it therefore satisfies both Norwegian and UK building regulations. The stair tool in Revit makes controlling these parameters quite simple, compared to having to calculate and draw this manually.
A stair can be a thing of beauty and is often a personal signature of
the architect! I have not yet acquired control over the design of the actual stair beyond the basic parameters, so much more practice with this is required before I can create a custom design.
This was also another useful place to create a cutaway view. Initially, whilst drawing the walls I marked out the position of the chimney with a column stretching through all floors. This enabled me to place the bedroom walls in a non conflicting position before returning to this in more detail.
I later found a fireplace and pre-fabricated chimney on BIMobject, and positioned these on the model. The pipe connecting them was something of a quick fix, a simple cylinder created in Component>Model in-place and assigned the same colour as the Nordpeis unit (as displayed in their marketing material, but not included in the model). This extends to roof level to connect to the chimney.
On the floors above, I boxed in the pipe with a wall casing. I looked at various chimney stack options, for example here by Schiedel to get an idea of various construction options. I was satisfied with this as a simplification ahead of any further detailed design which may be required later at a construction phase.
I wanted to see how components worked in Revit, and decided to try and use the included Revit kitchen units to create a functional layout. I tried drawing this same kitchen initially with units found on BIMobject from Ballingslöv. However, the models weren’t as easily configurable as I hoped and looked untidy in plan.
It seems the advantage with the built-in components is that they are designed especially for use with Revit and have a readily refined appearance in plan and elevation. They are a little limited in style and appearance, but serve to illustrate a purpose or intention.
For the kitchen design, I installed an additional wall where appliances and tall cupboards could be recessed. Also, an kitchen island works well in this space with a breakfast bar.
These first two weeks of using Revit has brought me to a point where I am able to construct a reasonable looking building model. From here I will want to further my knowledge of detailed construction, and be able to refine and control the model to give useful data. I am also very curious about how to create custom geometries or better integrate manufacturers products. I am particularly curious about how it might be implemented in the renovation of heritage projects, and have located some relevant tutorial series to spend the time in Corona-lockdown with!
For now, here are some images from my first attempts at drawing in Revit:
Using Autodesk Inventor, the task is to design a casing for the Raspberry Pi unit, as shown below.
The brief for the task is to meet the following criteria:
- The Raspberry Pi should be securely held in place by the casing.
- The casing should include the Raspberry Pi logo as a decal, or preferably embossed in a fitting place.
- The following interfaces/ports must be easily accessible from the outside of the case:
- 4 x USB2 Ports
- 10/ LAN Port
- 3.5mm 4-pole Composite Video and Audio Output Jack
- Full Size HDMI Video Output
- Micro USB Power Input
- MicroSD card Slot
- The design should be suitable to be used in a modern day living room, aimed at reaching the growing Home Theatre PC market.
- The design should incorporate a passive cooling system
- The design should have a two part casing where there is a feature that guides parts together for easy assembly.
The first task was to review a selection of some of the wide range of cases already available. I collected some links together on a Pinterest board: https://pin.it/7wh1LK5
One that particularly caught my attention was this elegant solution, by Flirc. The aluminium body also acts as a heat sink. The assembly comprises only three main parts, plus 4 screws and a heat transfer pad.
This commercially available ‘Quatro’unit also fits the brief of fitting into a modern living room setting. It is a slick 4 part plastic square casing, with allowance for extra camera or cooling additions to the Pi as desired.
IKEA’s Bestå series is typical of the modern solution for TV and media storage. A fully configurable system allows variations in fronts, styles and finishes, but is primarily based upon retaining clean lines and simple geometry.
I developed a simple design, based on the idea of matching the unit to a clean minimalist, but functional style that would fit in a modern living room.
- Minimalistic approach – pure geometry /hidden detailing
- Two part casing with rubber feet
- Heat sink incorporated into lid design
- Machined Aluminium lid
- Plastic or aluminium body
Contrary to the exercises with the model train set, I began in the assembly and drew most of the components from there. I started with the extrusion highlighted below, where the casing meets the attachment holes in the Pi circuit board.
The design continued outwards from here. The overall dimensions of the case are linked by constraints to the outer dimensions of the Raspberry Pi unit:
I learned here that the initial concept sketch was an important aspect of the design. Realising a detailed model is difficult without having a developed concept already in place. As with many CAD packages I have previously worked with, it seems Inventor is a powerful tool in developing a design in later stages rather than the initial phases. The amount of detail possible here can I think be a distraction from developing simple ideas, which is where simple hand drawn sketches will always excel.
As with the Flirc design, I have tried to incorporate the heat sink within the casing. My thought for the material of this element is machined aluminium. It is intended that heat transferred from the CPU could be dispersed through a heat pad and up to the lid through a solid element. The lid features grooves and the Raspberry Pi logo, to maximise surface area and cooling effect.
The joining point between the two case elements is an area requiring some further development. The difficulty is how to retain a minimalist outer appearance, but incorporating a latch system where the lid will remain in place. Otherwise, edge profiles as detailed in the brief were achieved as seen here.
Inevitably, after only 2 weeks of using a new piece of software there are some areas to develop. I see for example in this model, that I have built the assembly on its side. When later trying to add a ground plane for the visualisations, I realised the problems that now causes! Left in the viewcube, the top view should in fact be the front. From the way I have built this model, there doesn’t appear to be a quick fix to this. This is something I will keep in mind for the next project. Interestingly though, the issue fixed itself when exported to the dwfx viewer (see below).
Scale is also something to be wary of. I realised late that the dimensions of this box are much smaller than I had initially imagined. There is a chamfer of 0.1mm on the edges, but whether this is achievable or not with this process and material has not been thoroughly considered. With the facility of infinite zoom, it is easy to lose perception of how large or small something on the screen actually is in reality.
The project was a success in terms of learning how to develop an assembly using Autodesk Inventor. Given the timescale of 1.5 days for design and modelling, it is not surprising that this particular design is not as refined as the examples examined at the beginning. If this project was being taken further, there are things I would change. For example, the profiling on the lid came sponaneously through experimenting with the software, rather than being a considered visual aesthetic. The lack of a latch system has already been mentioned. Prototyping would no doubt also reveal some issues with precision in manufacture. However, this was always an exercise in 3D modelling and illustration and to that extent it is complete. To finish, I experimented with exporting to dwfx and to png with Inventor’s export, rendering and section view options:
Whilst Norway has been shutting down and enforcing isolation due to the spread of Covid-19 this week, it is somewhat business as usual for those of us already studying online programmes. Whilst the rest of the world is switching to digital platforms, these strange times we are experiencing have given me even more excuse to get stuck into a new piece of software: Autodesk Inventor.
Here are the result of my first efforts, following an online tutorial from the Noroff Technical Design 2D/3D CAD course to get a hang of the basic principles using a toy train locomotive:
I can already see this is a very capable piece of software for product design and assemblies. There are many differences from 3D packages I have used previously, for example AutoCAD or 3DsMax. Splitting models into parts and assemblies for example, and also the use of work planes and sketching. Oddly, some of the terminology and methods differ from other Autodesk products despite obvious similarities. So there is much yet to develop a full understanding of; such as modelling and assembling with constraints. I look forward to this in the coming weeks of enforced isolation!
For me, one of the most intriguing areas of development in 3D modelling technology has been its recent use in digital heritage. Specifically, the use of digital scanning to create detailed records of ancient monuments and artefacts. After the disaster at Notre-Dame de Paris nearly one year ago, much was written about the existence of a detailed record of the building after a digital scan completed by pioneering researcher Andrew Tallon. That survey may prove invaluable in restoring this great work of architecture. Institutions such as CyArk are doing similar work documenting other important sites around the globe, safeguarding valuable treasures in a digital vault for future generations.
Photogrammetry received a quick mention in the Noroff course material, and I jumped at the chance to try it out at a very basic level. The concept is that an accurate 3D mesh or point cloud model can be created from a series of 2D digital photographs. The potential applications for this extend from digital heritage to complex topological surveying, scene creation for 3D animation and a great deal more.
Autodesk have a piece of software called ReCap, which is what I’ll be using to create a scanned model. This is compatible with other Autodesk products, meaning scanned models can be imported into the likes of AutoCAD or Revit. I’m eager to test out the possible applications for architectural design here, but thought to get started with a simple object first.
Below is a little rubber frog, a replica of the one owned and carried as a lucky charm by composer Edvard Grieg whilst performing (and incidentally was bought from the gift shop at his former home here in Bergen). I chose this because of its rough texture and organic shape, meaning it should be a suitable object to scan with a regular iPhone 6 camera (which has seen better days!). Smooth surfaces are apparently difficult to capture, whilst an object based on simple geometry rather defeats the point of wanting to create a scan in the first place. So Grieg’s frog it is.
These are the set of images I took on a window ledge (mostly to get good natural lighting on the object), which were then imported into ReCap photo (a side module to ReCap specifically to handle the photographing/stitching aspect).
It took a little while to stitch the images together, and things seemed to be moving slowly. I left the software with a ‘waiting in cue 1% complete’ message and came back a day later to find a completed model. I then had various options for patching up the model, and various formats for exporting including video/png.
The model itself came back with a few incomplete areas, and very much of the window ledge which I trimmed away to just the rectangle you see. The white areas on the frog shows how important it is to get photographs from all possible angles. I made that difficult by having it on the window ledge, and therefore the back and underside of the frog are not shown accurately. The photo points are shown in the screenshot below by the triangles, mostly from above the model.
I had previously read about others having problems due to moving the object in relation to the background. For the scan to work, either the object or the camera has to remain still. In my case, having the window and the ledge there made getting a full orbit difficult. Trying again with the object elevated on a small platform of some sort would perhaps produce a better result.
For a first attempt, this was quite an exciting experiment! Room for improvement of course, but the applications for this technology on even a very low budget scale could be very useful. If one begins to look at the scan below as a potential building form, the potential uses for application in design process become more apparent. It allows a designer to get hands on with a prototype or physical model, and then scan the geometries to reproduce them digitally. Until this, the exchange of CAD data to the real world had mostly been one way but now it seems there is a very accessible, affordable and simple way to input complex geometries back into the digital modelspace and become part of the workflow.
The task is to create ‘accurate technical drawings’ using 3D modelling tools for the following coffee table, using the following drawing as a source.
As an extra challenge, the ISO 4762 screw may also be modelled and included in the drawings.
As a second additional challenge, tolerances are to be chosen and applied that will ensure that the parts fit together.
Venturing into three dimensions
This project is the first assignment drawn in 3D on the course, and serves as an introduction to working in the 3D environment. Though I have some previous experience with 3D modelling, I have discovered a much upgraded 3D-interface in this latest version of AutoCAD which makes this process more intuitive and efficient. I have also invested in a 3Dconnexion Spacemouse which makes navigation much more effective, though that took some practice before it became useful! This video shows me getting to grips with that!
The most important difference I have been learning these past two weeks is how to go back and make adjustments to a model in AutoCAD. When I was using earlier versions of AutoCAD in the profession around 2008/09, I remember this was a problem in the 3D environment. Instead, we opted for SketchUp because it was much simpler to work with. Now I see that AutoCAD has incorporated and developed SketchUp’s main modelling feature; the Presspull tool. Along with the Ribbon toolbar and a more intuitive UCS tool, this makes 3D modelling in AutoCAD much less frustrating and far more practical than I experienced before.
The ease of which 2D drawings can be created from the model was an exciting revelation! This is an area I think would benefit from a lot of further practice with more complex models, especially with layer management. However, I am very pleased after this exercise with how relatively simple it was to extract a useful drawing and then go back and edit something when I realised it was wrong.
Solids, Booleans and a fillet
This was a well chosen task to introduce us to a range of different basic modelling techniques using solids. I was able to familiarise myself with all of the basic elements and boolean operations, as well as using fillet edge and shell commands. Here is an example of how various geometries make up the table:
The taming of the screw
Modelling a screw is one of the more complex common tasks in AutoCAD, according to this web tutorial I stumbled upon. I was keen to try, and to try and work out how to break down the geometry.
Interpreting the screw data can be challenging, I have found. I based what I could on the tables given, and the remainder from the tables in Maskintegning (specifically thread geometry). I am not certain it is 100% accurate, especially the beginning and ends of the thread. However, as a first attempt at modelling a screw, I am satisfied that I have something that looks correct with the most critical of dimensions in place.
In our brief that there wasn’t any specification of material. The drawing also lacks information about the thickness of the table top, or whether for example the handle element is a solid element and perhaps part of the same moulding?
I have walked past a table that closely resembles our design many times at IKEA (I work there part-time, which also provides useful opportunity to learn!). I took the liberty of researching the ‘YPPERLIG’ table (left) and seeing how this version was assembled, and most importantly what it was made from. In this instance, the table top is powder-coated steel and the legs are made from solid birch. There is no handle element here but on another similar product, hollow steel tube is used. I have used this to fill in some of the missing information from the supplied drawing, for example to assume a thickness of 2mm for both the table top and tube handle.
I have not indicated any type of fastening for the handle. I would anticipate that if this was the solution, two additional screws would be required here and that this would be included as a separate part. Since this is a deviation from the brief and the original design, this is omitted but it would be a case to bring up as part of the review of these drawings in a real life scenario.
Upon observation, another critical difference from the IKEA table was the geometry of the legs. On the drawing we were provided with, it appears there was a square section coming off the cylinder, for which I have drawn with an assumed width of 4mm. With the ‘YPPERLIG’ table, I see the legs themselves are simple cylinders. I assume this is to keep manufacturing costs low, using standard stock raw material with minimal processes (a cylindrical birch post in this instance). In our version, I have drawn the asymmetric posts and guess that along with the spherical tops and bottoms, these geometries are there to increase the difficulty of the challenge as a modelling exercise! I would be curious though as to how that would be achieved in production and what effect that would have on production costs.
The second additional task was to include details of tolerances. After reading the relevant chapters of Maskintegning and Technical Drawing with Engineering Graphics I was overwhelmed with possibilities. Without knowing so much in detail about the production and materials, or the effect of those tolerances on the assembly, I opted for a simple expression of limits with an allowance of 2mm on all parts. On all drawings, the following note was included:
All tolerances ± 1mm unless otherwise stated.
Mating parts have been given a different set of limits, still with an allowance of 2mm but so that they will not exceed a maximum that would prevent them from fitting together. For example:
Both of the 30mm slots on the upper and lower legs have a stated maximum of 32mm and minimum of 30mm. If the dimension was stated only at 30mm, there is a risk the parts could be accepted at 29mm each. This would mean the parts were too tight to fit, therefore the limits have been adjusted on these dimensions and all others that apply to connecting parts.
See model in Autodesk Viewer here:
The task is to produce assembly and detail drawings for the Garden Looper (or Shears as I would call them?!) product from the source drawing below.
Untangling the dimension spaghetti
My first reflection to this task was; what a horrible source drawing! Handle 1 was perhaps the most difficult to interpret and understand. A potential error in either the drawing itself, or of my interpretation of it is described later below, see ‘Collision?’.
Annotation lines complicate the clarity of this isometric drawing, which also appears to be shown as a wireframe from a reverse angle. The size of the drawing also makes the information difficult to read. However, I have already experienced that having inadequate drawings to work from is not an untypical situation in a real working scenario and often the reason further drawings are required in the first place.
Some of the dimensioning techniques here also had me reaching for the textbooks, requiring further revision/rereading from earlier lessons. The use of figures ‘2X 28°’ and ‘3X R2’ were new to me, and took a little deciphering before I figured out what they were describing. 2X in this instance saves dimensioning both of the 2 repeated angles of 28°, whereas the 3 gear teeth each had a radius of 2mm to their outermost arcs. This led to the construction below:
Once I had got my head around the dimensioning techniques, the rest of the detailed drawings were straightforward enough to complete. My initial thought was to combine each of them into the assembly drawing as an Xref, which would have the advantage of being able to make adjustments in the original drawings without having to do that again in the assembly drawing.
Below is a screenshot from an instance where I had ‘show frames’ activated, in an attempt to block out parts of the drawings which were not required. In practice, this did not get me the results I had hoped for. There were too many parts of these drawings which needed to be masked out, and in the end I couldn’t find a more effective way to achieve the drawing I required than to bind and explode the XRef into the drawing.
In retrospect, using blocks may have been a wiser choice in this instance, rather than entire dwgs. There would still be the issue of how to ‘mask out’ areas of those parts which sit behind others in the assembly. This is an area which will require further practice and refinement.
The specification ‘M6 X12’ was something I also had to do some reading up on before being able to draw the correct screw, understanding that all screws and Metric threads follow standards set by the ISO. With help from the Tingstad catalogue I found the required dimensions for the specified screw and was able to draw this accurately as a block within my assembly drawing.
With the understanding that the screw length here is 12mm, and the cutting blade having a thickness of 2mm, I determined that 10mm was the required depth for the two holes in each of the handles.
There is one major issue with the drawings I have produced. Having attempted to redraw that which I have been sent, I can identify a collision in the assembly drawing (01) . That is that the cutting blade appears to overlap with Handle 1. I have rechecked my drawing of the original material and do not see where this error has occurred according to my interpretation of the original drawing. However, that drawing lacks clarity in describing this particular part of the assembly. In reality, this is an issue I would discuss with the designer before sending anything to production.
The red arrows above point to where there appears to be a collision. It is not clear whether this is an error in interpreting the information provided, or an issue with the original design.
Assembly drawing doubts and Isometric pains
I had my doubts over the usefulness of the assembly drawing (01) after it was completed. I struggle to see a section line that adequately describes all the elements of assembly. From the start I imagined that an exploded assembly drawing would better it, but since the Isometric drawing was listed as an additional challenge to the task I began with a section as the initial starting point. In the end, the additional challenge of the isometric drawing was included as Drawing no.06, intended as an alternative to no.01.
These isometric views were difficult to construct! The biggest challenges were drawing the geometries based upon angles as described earlier, but in Isometric view. Since this was not possible to recreate within isometric circles, construction lines from the original drawing were copied in and rotated by 30 degrees where necessary. However, this was a time consuming process.
Ideally, Handle no2 would have been drawn at its assembled angle. However, that would have made constructing the geometry of that part in isometric view extremely challenging and time consuming.
The biggest stumbling block for me here though was the elliptical shape of the handle. I found no satisfactory method of constructing that shape in Isometric view. The final result was an approximation, which I think for the purpose of this particular drawing doesn’t create any issues. However, I’d like to find out how to do this accurately or if it is even possible using AutoCAD 2D!
Even more difficult, is the very end of the grip where they curve in 2 planes. From this angle it is perhaps impossible to draw accurately where the end of the grip will in fact appear, so again it was improvised using the central outermost isocircle at the centre of the grip, and arcs connecting to it to represent the other outermost points.
The full document contained details for all parts, with an assembly drawing that looked something like this:
The goal of this assignment is to produce an updated floor plan for the log cabin below. The drawing will be updated with a wall construction in accordance with building regulation TEK10, using the appropriate wall thickness for the use of 200mm Glava insulation.
This will be the first architectural drawing produced on the course. The work will involve using many of the features learned in AutoCAD over the past 2 weeks, for example; dimensioning, use of layers and plot styles, and publishing to dwf in both A1 and A3 layout sizes.
This task presented me with some challenges, and due to other activities this week a time constraint which demanded a healthy efficiency of learning and working. Aside from demonstrating the use of the tools that have been taught over the past two weeks, there were also some technical difficulties to overcome in adapting the plan to a new building standard. For the first time on this course, I was also confronted with regulations and standards.
TEK10 was the set of regulations implemented in 2010 for the minimum requirements a building must fulfill in order to be legally erected in Norway. It has since been succeeded by TEK17, which was implemented in 2017 and are the regulations now in current use. Through documentation from SINTEF and Norsk Standard (NS), details for pre-approved solutions can be found to meet the regulations. Manufacturers such as Glava Insulation also provide a range of products designed to meet those regulations.
The first difficulty I ran into with this assignment was the precision required to draw at. I concluded that the outer wall needed to be drawn at 198mm (though in reality it would be much thicker with cladding, damp-proof membrane and inner finishes). The original log construction was 120mm. The assignment also stipulated that an inner wall thickness of 98 mm should be used. So here the outer walls would be 78mm thicker, whilst inner walls would be 22mm thinner.
Additionally, it said lengths are to be drawn at 10mm increments. However, with the walls being 198 and 98mm, I found this a challenge to maintain throughout the entirety. Another difficultly I found was in considering which way to offset the new wall dimension in order to retain a similar room area. I set the original drawing under my AutoCAD drawing and scaled it as close as I could to the correct size. That way I was able to see the wall plan and dimensions of the original, and keep as close as possible to that.
From the screenshot image during production, you can see the exterior wall was drawn based on the interior dimension, so that the internal floor area would not be diminished. I encountered a dilemma when drawing the stores, not knowing whether to insulate them or not. I would need more information about the building to make a decision on this. Presumably they are outdoor stores, for ski equipment perhaps and maybe not in need of insulation. The external wall here might though be load-bearing, so it is still drawn as an exterior wall. So too is the wall on the interior side, presuming that will instead require insulation. In conclusion, this resolution requires more information which is not actually relevant to this task as a drawing exercise, so it is left unresolved.
The next problem I encountered was that the provided door blocks had some inaccuracies, which led to some distorted dimensions. Perhaps this was done deliberately to test us! In any case, I found information through the TED forum of the correct door sizes listed from Swedoor, including frame and opening and corrected the blocks in my drawing to the correct size.
I was uncertain in the end whether I should perhaps have been drawing to a 10mm tolerance, with wall thicknesses of 200 and 100mm. However, the door blocks provided were to a 1mm tolerance. In the end, I drew to 1mm and tried to keep wall positions to 10mm increments, with the exception of two windows which were positioned centrally within a wall.
Some time earlier in the week was spent setting up templates for the forthcoming drawings, with various layers and plot settings. Again, I understand there are standards in Norway for how these are set up and I understand how the need for this has grown as information is shared more and more digitally. I have learned this week to be more disciplined in their usage, and to try and avoid creating additional unnecessary layers. The PURGE command was also used to remove some that crept in during importation of some blocks.
Annotative dimensioning is an aspect of AutoCAD which has really developed in recent versions. The ability to create one dimension with the ability to change its scale according to the view it will display in seems to be a very advantageous way of working. Therefore I have chosen to adopt this method, but am aware of the others using layout space. I am now also aware that there are standards for text heights and linetypes, again defined by NS.
The next challenge was finding or drawing suitable blocks to represent furnishings as part of the additional task, which were also in accordance with NS. I downloaded some blocks from an online CAD block library. Some of these are based on actual furniture items, such as sofas and coffee tables. The beds and dining table are generic sizes. I also searched for and identified NS kitchen symbols for ovens, refrigerator and dishwasher. This is something new to me in this country and perhaps not standardised in the UK.
Part two of the Week 2 assignment was to draw the same oven as an isometric projection.
Isometric drawing in AutoCAD is something I had little experience of before last week, so this task has been a little revolutionary for me. Isometric drawing in itself is something I have tried by hand, but not in CAD. Once I understood better how to control the polar tracking (F10) and Isoplane settings (F5), this was a straightforward task. I also found out how to draw an ellipse to appear as a circle in Isometric mode (Isocircle), which was a pleasing moment!
To my mind, the simple finished line drawing bared resemblance to the front cover of an IKEA assembly guide (to me at least). The style of these manuals are somewhat iconic and familiar, and have provided varying degrees of helpfulness to most of us at some point!
Having completed the task in good time, as a sidestep I decided to analyse how their house-style was composed. In typical Scandinavian style; bold fonts and outlines, lots of white space around the drawing, simple legibility. This is a functional graphic style, and not in any way a production drawing, but nevertheless one based upon simple line drawings.
Just for fun, I tried to replicate it here using AutoCAD, exporting to PDF and converting to .png to be uploaded here. Apart from the design of the hob, the biggest difference is that they are using perspective drawings presumably extracted from a 3D model. I have been using Isometric projection, a flat drawing based upon 30 degree angles. It also lacks scale, since it is not a production drawing. The conclusion: AutoCAD has some usefulness in graphic illustration, not just CAD production.
I also tried adding some ‘shadow’ behind the control dials, as per the Ikea drawing. For this, I have dabbled a little with the text style and line weight settings, which I hope to getter a deeper understanding of later. Finishing touch of course, a slightly cryptic and amusing Swedish name!
Week 2 of the course, and the focus switches to reporting what I have been learning by means of a first drawing assignment.
An orthographic drawing of a stove unit has proved to be a simple enough task, utilising tools and commands in AutoCAD which have become refamiliarised through this process. There were however a few surprises along the way.
Hooray for array
To create the 6 identical switches on the oven, I chose to use the Array tool. As per the assignment, the switch itself was a simple block. New to me though was that the Array command is much enhanced and can be adjusted afterwards too. In this instance, it was a perfect shortcut to copying out the block 5 times, and one which allows me to make changes afterwards if needed.
When it came to presenting the drawing on a title sheet I had to learn more about the Enhanced Attribute Editor in order to add the details of the drawing to the title block. A new feature (to me at least) which is incredibly useful for keeping blocks and drawing templates very standardised and clean. The editor itself was surprisingly intuitive and simple to use, contrary to my previous experience with AutoCAD!
Autodesk Design Review
I have to admit, I’m yet to fully appreciate what this extra piece of software is for and what it can do. I have though managed to export my finished drawing as a .dwf and open it up in Autodesk’s Design Review. A few things here that I wouldn’t be satisfied with in a professional context; lack of line-weight or plot settings, and a lack of dimensioning. I hope this will be covered in coming lessons, and which I am in need of time to refresh upon.
The first stage of the assignment was completed within a couple of hours, including researching these new features and report writing. This was in addition to the time spent on the various helpful daily tasks provided. With time to spare, and more to learn; onto the additional task!